Ix

Figure 10.16. Smectite clay structure with the ideal talc formula MgaSUOiofOH^. (From R. Grim, Clay Mineralogy, McGraw-Hill, New York, 1968.)

layer are perfectly flat, with two-thirds of the oxygens in the two centrally located planes replaced by hydroxyl (OH) groups. All the tetrahedral sites are occupied by Si, and all the octahedral sites are occupied by Mg, forming the coordination groups Si04 and MgOé with the structures sketched in Figs. 10.17a and 10.17b, respectively.

The clays to be discussed here are of the montmorillonite class, and the discussion will center around die particular clay called saponite, in which some aluminum (Al3+) replaces silicon in tetrahedral sites, and some divalent iron (Fe2+) replaces magnesium in octahedral sites, corresponding to the typical formula tNa63Ca00!5][Mg2»F^+][Si36Al04}O,0(OH)2 ■ 4H20. Each layer is 0.94nm thick,

Figure 10.17. Structures of (a) Si04 showing the silicon atom centered in a regular tetrahedron of oxygens and (b) MgOe showing the magnesium atom centered in a regular octahedron of oxygens.

Figure 10.17. Structures of (a) Si04 showing the silicon atom centered in a regular tetrahedron of oxygens and (b) MgOe showing the magnesium atom centered in a regular octahedron of oxygens.

with a top and bottom surface area of 660m2/g. It has been found possible to intercalate or place between the layers bulky ions that form pillars that hold the layers apart, as shown in Fig. 10.15c, and thereby provide a system of spaces or pores where various small molecules can reside. The dimensions of the pores produced by this layering process are in the low-nanometer range. These materials are called pillared inorganic layered compounds (PILCs).

The pillaring is often carried otó with the aid of the positively charged All3 Keggin ion, which has the formula [A11304(0H)24(H20)12]7+, or alternately {A104[Al(0H)2H2O],2]7+. The aluminum atom groups depicted by circles (i.e., spheres) are arranged in the close-packed structure sketched in Fig. 10.18, which depicts a midplane containing six visible and one centrally located but obscured atom group, plus three such atom groups above them. The three below are not shown. In this ion the centrally located aluminum is tetrahedrally coordinated, and the surrounding 12 aluminums are octahedral fy coordinated, as delineated in Fig. 10.19. The Keggin ion is sometimes prepared in A1C13 solutions by transforming pairs of the trivalent hexaaquo ions A1(H20)¿+ to dimers A12(0H)2(H20)^, which are subsequently coalesced to form the Keggin ion. The presence of this ion can be unambiguously established by an 27 Al nuclear magnetic resonance (NMR) measurement because its central tetrahedrally coordinated aluminum ion produces a narrow NMR line that is chemically shifted by 62.8 ppm relative to hexaaquo aluminum Al(H20)g+. The twelve octahedrally coordinated aluminums of the Keggin ion do not appear on the NMR spectrum because they are rapidly quadrupolar relaxed by the aluminum nucleus, which has nuclear spin / = §, and hence their NMR spectral lines are broadened beyond detection.

One way to carry out the pillaring process is to suspend the clay in a solution containing Keggin ions at a controlled pH (i.e., acidity) and a controlled OH: Al ratio. Some of the initial charge of +7 on the ion is compensated for when they bond with the silicate layers to form the pillars. The pillars themselves may be considered as cylinders with a diameter of 1.1 tun. There are about 6.5 saponite unit cells with the oxygen content O20(OH)4 per pillar. Taking the value A = 3.15 nm2 for die area

Figure 10.18. A dose-packed Al13 cluster showing one atom (almost obscured) in the center, six atoms in the plane around it, and three atoms at close-packed sites above it. Three more atoms, not shown, are at close-packed sites in the plane below.

Figure 10.19. Sketch of the structure of the (AKD4[AI{OH)2 H20]12 )7+ Keggin ion with one tetrahedral group AI04 in the center position of Fig. 10.18, surrounded by 12 AIOe octahedra at the remaining sites of Fig. 10.18, where the oxygens O, of the octahedra belong to hydroxyl groups OH, or to water molecules HzO. [From A. Clearfield, in Moser (1996), Chapter 14, p. 348.]

Figure 10.19. Sketch of the structure of the (AKD4[AI{OH)2 H20]12 )7+ Keggin ion with one tetrahedral group AI04 in the center position of Fig. 10.18, surrounded by 12 AIOe octahedra at the remaining sites of Fig. 10.18, where the oxygens O, of the octahedra belong to hydroxyl groups OH, or to water molecules HzO. [From A. Clearfield, in Moser (1996), Chapter 14, p. 348.]

of the silicate layers in 6.5 unit cells, the distance between nearest-neighbor pillars can be estimated. If the pillars are assumed to form a square lattice, then the spacing between the center points of nearest-neighbor pillars is (A)XI1 — 1.77 nm, and if they are arranged on a regular triangular or hexagonal lattice, then their separation is = 1.90nm Taking an average of these two values, one obtains a free space of about 0.74 nm between pillars. Experimental measurements indicate that the resulting pillared clay has a basal spacing of =1.85 nm, a surface area of = 250m2/g, and a pore volume of 5i0.2 cm3/g.

One important aspect of pillared clays that contributes to their catalytic properties is the presence of acid sites, which can be of the Lewis or Brensted types, as explained in the previous section. When a pillared clay is heated, the water and hydroxyl groups split out protons to balance the negative charges of the layers as the pillars approach electrical neutrality, and this generates considerable Brensted acidity. Lewis acid sites are generated on the layers by defect formation, and on the pillars by dehydroxylation, or the removal of OH groups. To confirm the presence of these sites on the PILC surface, the heterocyclic ring compound pyridine (C5H5N) was adsorbed and an infrared spectrum was recorded. This spectrum exhibited a strong IR band at 1453 cm-1 arising from Lewis acid sites, and a weaker band at 1550 cm-1 produced by Brensted sites.

The discussion until now has centered around the clay saponite pillared by the Keggin ion. Other types of montmorillonite clay materials have been used, and other metal oxide polymers have served as pillars. Examples of nonalumina pillars are based on the proposed zirconium ion Zrlg04(0H)36(S04)14, the bivalent titanium ion [Ti(CH3COO)6 4(OH)0 4C1,2JC1 • 11H20, hexavalent chromium octahedra forming the ions [Cr4(0H)6(H20)u]6+ and [Cr4O(OH)6(H2O)10]5+, an alumina-silica A1203—Si02 combination, and silica Si02 supplemented by some titania Ti02. The availability of these nonaluminum pillars, which differ in their dimensions, can provide catalysts with a wide range of pore sizes. These catalysts have been studied for their capability in carrying out various chemical reactions, such as cracking, in which hydrocarbons or other molecules are broken up and their fragments are recombined into desirable product molecules. An example is crude oil and gas cracking to produce gasoline. One of the liabilities of these pillared catalysts is their tendency toward coke formation whereby the surface becomes coated with carbon, and acid sites become deactivated or unable to function.

10.2.5. Colloids

Nanosized particles of metals are ordinarily insoluble in inorganic or organic solvents, but if they can be prepared in colloidal form, they can function more readily as catalysts. A colloid is a suspension of particles in the range from 1 nm to 1 pm (i.e., lOOOnm) in size, larger than most ordinary molecules, but still too small to be seen by die naked eye. Many colloidal particles can, however, be detected by the way they scatter light, such as dust particles in air. These particles are in a state of constant random movement called Brownian motion arising from collisions with solvent molecules, which themselves are in motion. Particles are kept in suspension by repulsive electrostatic forces between them. The addition of salt to a colloid can weaken these forces and cause the suspended particles to gather into aggregates, and eventually they collect as a sediment at the bottom of die solvent. This process of the settling out of a colloid is called ftocculation. Some of the colloidal systems to be discussed are colloidal dispersions of insoluble materials (e.g., nanoparticles) in organic liquids, and these are called organosols. Analogous colloidal dispersions in water are called hydrosols.

In Section 2.1.3 we discussed the formation of face-centered cubic nanoparticles such as Au^j with structural magic numbers of atoms, in this case 55. This nanoparticle has been ligand-stabilized in die form Au^^PPh^) to make it less reactive, and hence more stable. This sturdiness is brought about by adding atomic or organic groups between the atoms of the cluster, or on their surfaces. These FCC metallic nanoparticles can be stabilized as colloids by the use of surfactants, which can operate, for example, by lowering the surface tension. The ring compounds tetrahydrofuran (THF) and tetrahydrothiophene, with structures sketched in Fig. 10.20, have been used to stabilize metallic nanoparticles as colloids. Figure 10.21 shows a Tin nanocluster coordinated with the oxygen atoms of six THF molecules in an octahedral configuration. In this cluster the Ti—Ti distance (0.2804 nm) is slightiy less than that (0.289 nm) in the bulk metal.

A way to obtain colloidal dispersions in organic liquids is to stabilize a metallic core using a lipophilic surfactant tetraalkylammonium halide NE^X, where X is a halogen such as chlorine (CI) or bromine (Br), and R represents the alkyl group

0 0

Post a comment